Background:
Co-testing (HPV DNA testing plus liquid-based cytology [LBC] for all screened women) is still used in some parts of China, but it may substantially increase cytology and referral workload—an important concern for county programmes with limited workforce and specialist capacity. 
Objectives:
To quantify the incremental detection benefit and the additional resource burden of co-testing compared with HPV-based and cytology-only strategies in routine primary-level cervical cancer screening services.
Methods:
We analysed a real-world screening cohort of 33,387 women aged 35–64 years from four primary care sites in central and western China where all participants received both HPV testing (12-type panel with HPV16/18 genotyping) and LBC at the same visit. Using the observed paired results, we deterministically reconstructed four strategies within the same population: (1) co-testing (refer if HPV16/18 positive or LBC ≥ASC-US), (2) HPV primary screening with LBC triage (refer HPV16/18; triage other hrHPV positives with LBC, refer if LBC ≥ASC-US), (3) HPV-only (refer HPV16/18; other hrHPV positives scheduled for 1-year repeat, no LBC triage), and (4) LBC-only (refer if LBC ≥ASC-US). Outcomes were estimated per 1,000 women screened, including CIN2+ detection, colposcopy referrals, and cytology workload (LBC slides). We additionally calculated positive predictive value (PPV) for CIN2+ among women completing colposcopy and the number needed to refer (NNR).
Results:
CIN2+ detection per 1,000 screened was 6.7 with co-testing, 6.5 with HPV primary + LBC triage, 4.3 with HPV-only, and 4.9 with LBC-only. Co-testing produced little incremental detection versus HPV primary + LBC triage (+0.15 CIN2+ per 1,000; 95% CI −1.08 to 1.38) but required substantially more resources: +33.1 additional colposcopy referrals per 1,000 (95% CI 29.50 to 36.70) and +888.8 additional LBC slides per 1,000 (95% CI 885.4 to 892.1). Efficiency measures were consistent with these trade-offs: PPV for CIN2+ was lower for co-testing (10.7%) than for HPV primary + LBC triage (18.0%), and NNR was higher (11.5 vs 6.7). LBC-only detected fewer CIN2+ than triage (−1.68 per 1,000) while still generating more referrals (+16.95 per 1,000). HPV-only reduced referrals (−20.07 per 1,000) and eliminated LBC workload but detected fewer CIN2+ (−2.28 per 1,000) than triage.
Conclusions/Implications for practice or policy:
In health-resource-limited county programmes, HPV primary screening with cytology triage offers the most favourable balance between high-grade lesion detection and downstream workload. Routine co-testing adds minimal detection benefit but markedly increases cytology and colposcopy burden, supporting policies that pivot away from co-testing toward HPV-based pathways tailored to local capacity.

study flow